Decision tree classification of land cover from remotely sensed data

Decision tree classification algorithms have significant potential for land cover mapping problems and have not been tested in detail by the remote sensing community relative to more conventional pattern recognition techniques such as maximum likelihood classification. In this paper, we present several types of decision tree classification algorithms arid evaluate them on three different remote sensing data sets. The decision tree classification algorithms tested include an univariate decision tree, a multivariate decision tree, and a hybrid decision tree capable of including several different types of classification algorithms within a single decision tree structure. Classification accuracies produced by each of these decision tree algorithms are compared with both maximum likelihood and linear discriminant function classifiers. Results from this analysis show that the decision tree algorithms consistently outperform the maximum likelihood and linear discriminant function classifiers in regard to classf — cation accuracy. In particular, the hybrid tree consistently produced the highest classification accuracies for the data sets tested. More generally, the results from this work show that decision trees have several advantages for remote sensing applications by virtue of their relatively simple, explicit, and intuitive classification structure. Further, decision tree algorithms are strictly nonparametric and, therefore, make no assumptions regarding the distribution of input data, and are flexible and robust with respect to nonlinear and noisy relations among input features and class labels.

A nonlinear model for assessing multiple probabilistic risks: A case study in South five-island of Changdao National Nature Reserve in China

Several methods for estimating the potential impacts caused by multiple probabilistic risks have been suggested. These existing methods mostly rely on the weight sum algorithm to address the need for integrated risk assessment. This paper develops a nonlinear model to perform such an assessment. The joint probability algorithm has been applied to the model development. An application of the developed model in South five-island of Changdao National Nature Reserve, China, combining remote sensing data and a GIS technique, provides a reasonable risk assessment. Based on the case study, we discuss the feasibility of the model. We propose that the model has the potential for use in identifying the regional primary stressor, investigating the most vulnerable habitat, and assessing the integrated impact of multiple stressors. (C) 2006 Elsevier Ltd. All rights reserved.

Composition, Abundance, and Diversity of the Peracarida on Different Vegetation Types in the Qi'ao-Dan'gan Island Mangrove Nature Reserve on Qi'ao Island in the Pearl River Estuary, China

Chang-Fu Wang, Xian-Qiu Ren, and Run-Lin Xu (2010) Composition, abundance, and diversity of the Peracarida on different vegetation types in the Qi'ao-Dan'gan Island Mangrove Nature Reserve on Qi'ao Island in the Pearl River estuary, China. Zoological Studies 49(5): 608-615. Almost nothing is known about the Peracarida in the Pearl River estuary. This is the 1st report to study the composition, abundance, and diversity of the Peracarida in the Qi'ao-Dan'gan I. Mangrove Nature Reserve on Qi'ao I., in the Pearl River estuary, southern China. Bimonthly samplings were carried out in 3 representative vegetation types (mangrove arbor, emergent plants, and seaweed) for 2 yr. Using a Peterson grab, 1940 individuals (id.) were collected in total, including 11 species of 6 genera, 5 families, and 3 orders (Amphipoda, Isopoda, and Tanaidacean). Discapseudes mackiei Bamber 1997 was the dominant species with the highest density of 1,432 incl./m(2). The effect of temperature on the abundance of Peracarida was significant (p < 0.01), and the optimum temperature was 22-23 degrees C in both the mangrove arbor and seaweed. The results showed that the abundance of the Peracarida was higher in the mangrove arbor, while the diversity, especially Amphipoda diversity, was higher in the seaweed. In contrast, emergent plants provided no suitable habitats for the Peracarida. http://zoolstud.sinica.edu.tw/Journals/49.5/608.pdf

The contribution of trace elements from seawater to chimneys: a case study of the native sulfur chimneys in the sea area off Kueishantao, northeast of Taiwan Island

Hydrothermal fluid containing abundant matter erupts from seafloor, meets ambient cold seawater and forms chimneys. So the main matter origins of chimneys are seawater and matter which are taken by hydrothermal fluid from deep reservoir. However, because of seawater's little contribution to the forming of chimneys, it is usually covered by the abundant matter which is taken by hydrothermal fluid. Therefore, chimneys formed in ordinary deep seawater hydrothermal activity, containing complex elements, cannot be used to study the seawater's contribution to their formation. While the native sulfur chimneys, formed by hydrothermal activity near the sea area off Kueishantao, are single sulfur composition (over 99%), and within chimneys distinct layers are seen. Different layers were sampled for trace element determination, with Inductively Coupled Plasma Mass Spectrometry (ICP-MS). By analyzing the data, we consider C-layer (secondary inner-layer) as the framework layer of the chimney which formed early (Fig. 4), and its trace elements derive from hydrothermal fluid. While the trace elements within A, B, D layers have undergone later alteration. A, B layers are affected by seawater and D layer by hydrothermal fluid. The increase of trace elements of A and B layers was calculated using C layer as background. Based on the known typical volume of chimneys of the near sea area off Kueishantao, we calculated the volume of seawater that contributed trace element to chimneys formation to be about 6.37 x 10(4) L. This simple quantified estimate may help us better understand the seafloor hydrothermal activity and chimneys.

Study of Radiolaria in the surface sediments from the area east of Taiwan Island

The species and characteristics of Radiolaria in the surface sediments were systematcally investigated in the sea east of Taiwan Island. One hundred and seventy-eight species of Radiolaria (including 21 unidentified species) have been identified in the surface sediments, and they belong to 2 orders, 34 families and 101 genera. Among them there are 19 families, 70 genera, 134 species of Spumellaria and 15 families, 31 genera, 44 species of Nassellaria. Of the 178 species of Radiolaria, the individual number of Spumellaria amounts to 88.1% of the total individual number, and that of Nassellaria amounts to 11.9% of the total individual number. It is shown that most of the dominant species belong to the tropical and subtropical dominant species and are brought into the area mainly by the Kuroshio, and some affecting factors including the submarine topography, submarine sediments, upwelling current east of Taiwan Island and carbonate dissolution play a secondary role in forming the Radiolaria distributions.

Sources and preservation of organic matter in Plum Island salt marsh sediments (MA, USA): long-chain n-alkanes and stable carbon isotope compositions

Elemental (TOC, TN, C/N) and stable carbon isotopic (delta(13)C) compositions and n-alkane (nC(16-38)) concentrations were measured for Spartina alterniflora, a C-4 marsh grass, Typha latifolia, a C-3 marsh grass, and three sediment cores collected from middle and upper estuarine sites from the Plum Island salt marshes. Our results indicated that the organic matter preserved in the sediments was highly affected by the marsh plants that dominated the sampling sites. delta(13)C values of organic matter preserved in the upper fresh water site sediment were more negative (-23.0+/-0.3) as affected by the C-3 plants than the values of organic matter preserved in the sediments of middle (-18.9+/-0.8) and mud flat sites (-19.4+/-0.1) as influenced mainly by the C4 marsh plants. The distribution of n-alkanes measured in all sediments showed similar patterns as those determined in the marsh grasses S. alterniflora and T. latifolia, and nC(21) to nC(33) long-chain n-alkanes were the major compounds determined in all sediment samples. The strong odd-to-even carbon numbered n-alkane predominance was found in all three sediments and nC(29) was the most abundant homologue in all samples measured. Both delta(13)C compositions of organic matter and n-alkane distributions in these sediments indicate that the marsh plants could contribute significant amount of organic matter preserved in Plum Island salt marsh sediments. This suggests that salt marshes play an important role in the cycling of nutrients and organic carbon in the estuary and adjacent coastal waters. (C) 2003 Elsevier Ltd. All rights reserved.

REPORT ON SOME SPECIES OF PALAEMONIDAE (DECAPODA: CARIDEA) FROM CLIPPERTON ISLAND

The present paper reports six genera nine species of Palaemonidae collected from Clipperton Island, eastern tropical Pacific, including three new species of the pontoniine genus Chacella: C. bicornuta, C. quadricornuta, and C. sexicornuta. We provide a key to identify all the six known species of Chacella.